CMA Staining Analysis of Chromosomes in Citrus Relatives, Clymenia, Eremocitrus and Microcitrus

Masashi Yamamoto1*, Asad Asadi Abkenar2,3, Ryoji Matsumoto2, Tatsuya Kubo1 and Shigeto Tominaga1

1Faculty of Agriculture, Kagoshima University, Korimoto, Kagoshima 890–0065, Japan 2Faculty of Agriculture, Saga University, Honjo-cho, Saga 840–8502, Japan 3Department of Breeding and Plant Improvement, Iran Citrus Research Institute, Romsar, P. O. Box 189, Iran

Fluorochrome staining with chromomycin A3 (CMA) was used to characterize and compare the CMA banding patterns of chromosomes of Clymenia, Eremocitrus, and Microcitrus, which belong to true citrus fruit trees (Citrinae, Rutaceae). All species had 2n = 18 chromosomes. These chromosomes were classified into six types based on the number and position of CMA-positive bands; A: two telomeric and one proximal band, B: one telomeric and one proximal band, C: two telomeric bands, D: one telomeric band, E: without bands, and Dst: type D with a satellite chromosome. Each species possessed three or four types of chromosomes and unique CMA banding patterns. The CMA banding patterns were 2C + 8D + 8E in Cl. polyandra, 2C + 9D + 7E in E. glauca, 1C + 11D + 6E in M. australis, 1B + 2C + 10D + 5E in M. australasica, 8C + 7D + 2E + 1Dst in M. inodora, 2A + 14D + 2Dst in M. warburgiana, and 2C + 9D + 7E in Sydney hybrid. Chromosome configurations of Cl. polyandra, E. glauca, M. australis, and Sydney hybrid resembled each other. This may indicate a common ground of chromosome configuration in the true citrus fruit trees. On the other hand, variability in Microcitrus chromosomes was also demonstrated.

Key Words: chromomycin A3, karyotype, phylogenic relationship, true citrus fruit trees.

Introduction Fraction I protein (Handa et al., 1986), and DNA (Asadi Abkenar et al., 2004a; Federici et al., 1998; Nicolosi et The true citrus fruit trees group of the subtribe Citrinae al., 2000) analyses. (Rutaceae) includes six genera, i.e., Citrus, Fortunella, Chromosome analysis using guanine–cytosine (GC) Poncirus, Clymenia, Eremocitrus, and Microcitrus. The specific fluorochrome chromomycin A3 (CMA) has been first four genera and the last distributed mainly in Asia found to be useful for determining the phylogenic and Australia, respectively. Despite the wide distribution relationships of Citrus, Poncirus, and Fortunella (Befu of these genera, cross compatibility of several et al., 2000, 2001, 2002; Carvalho et al., 2005; Cornelio combinations within this group has been reported et al., 2003; Guerra, 1993; Kunitake et al., 2005; Miranda (Barrett, 1985; Iwamasa et al., 1988). Not only is Citrus et al., 1997; Yamamoto and Tominaga, 2003; Yamamoto one of the most important fruit trees of the world, but et al., 2005, 2007). These studies demonstrated the the other five genera are also important for industry existence of characteristic CMA banding patterns with because of the characteristics of the fruits as well as the a high level of diversity and heterozygosity in the utilization as rootstock and resistance to biotic and chromosomes of the above genera. The results also abiotic stresses. Phylogenic relationships of these genera demonstrated CMA banding patterns of important have been elucidated by isozyme (Herrero et al., 1996; species, which provide useful information on phylogenic Rahman and Nito, 1994a, b; Rahman et al., 1994), relationships among these genera and species; however, CMA banding analysis of the other three genera has not progressed, although Guerra et al. (2000) demonstrated Received; May 31, 2007. Accepted; August 2, 2007. This research was supported by a Grant-in-Aid (No. 18580028) from CMA banding patterns of one species each of the Japan Society for the Promotion of Science. Eremocitrus and Microcitrus. * Corresponding author (E-mail: [email protected]). In this study, we clarified the variability of CMA

24 J. Japan. Soc. Hort. Sci. 77 (1): 24–27. 2008. 25 chromosome banding patterns in several species a satellite chromosome (Fig. 1). Each species examined belonging to Clymenia, Eremocitrus, and Microcitrus in this study exhibited high chromosomal variability with and discussed their phylogenic relationships. characteristic banding patterns. A light CMA-positive Materials and Methods band was observed in the satellite position of five chromosomes in M. warburgiana (Fig. 2). That type of In this study, Clymenia polyandra, Eremocitrus chromosome was included as type D in this study. glauca, Microcitrus australis, M. australasica, The CMA banding patterns were 2C + 8D + 8E in M. inodora, M. warburgiana, and Sydney hybrid Cl. polyandra and 2C + 9D + 7E in E. glauca. In species (M. australis × M. australasica) were used (Table 1). The belonged to Microcitrus, 1C + 11D + 6E in M. australis, materials used in this study were conserved at the Faculty 1B + 2C + 10D + 5E in M. australasica, 8C + 7D + 2E + of Agriculture, Saga University, Japan. 1Dst in M. inodora, 2A + 14D + 2Dst in M. warburgiana, Young leaves about 3–5 mm long from adult trees and 2C + 9D + 7E in Sydney hybrid (Fig. 2, Table 1). were used as materials. Young leaves were excised, E. glauca and the Sydney hybrid showed identical immersed in 2 mM 8-hydroxyquinoline at 10℃ for 4 h CMA banding patterns. Chromosome configurations of in the dark, fixed in methanol-acetic acid (3 : 1), and Cl. polyandra, M. australis and the aforementioned two stored at −20℃. species (accessions) resembled each other; the numbers Enzymatic maceration and air drying were performed of types C, D, and E chromosomes were similar. as described by Fukui (1996) with minor modifications. M. australasica possessed types B, C, D, and E The young leaves were washed in distilled water to chromosomes. All these chromosomes are observed remove the fixative and macerated in an enzyme mix- commonly in Citrus, and types D and E chromosomes ture containing 2% Cellulase Onozuka RS, 1.5% are also predominant in Citrus and Poncirus (Befu et Macerozyme R200 (Yakult, Japan), 0.3% Pectolyase Y- al., 2000, 2001; Carvalho et al., 2005; Cornelio et al., 23 (Seishin Pharmaceutical Co., Ltd, Japan), and 1 mM 2003; Guerra, 1993; Miranda et al., 1997; Yamamoto EDTA, pH 4.2, at 37℃ for 45–60 min. and Tominaga, 2003; Yamamoto et al., 2005, 2007). Chromosomes were stained with 2% Giemsa solution However, the chromosome configuration of M. inodora (Merck Co., Germany) in 1/30 M phosphate buffer (pH 6.8) for 15 min, rinsed with distilled water, air dried, and then mounted with xylene. After confirmation of each chromosome position on the slide glass, the chromosomes were de-stained with 70% methanol. Chromosomes were also stained with 0.1 g·L−1 CMA according to Hizume (1991), and observed under a fluorescence microscope with a BV filter cassette. Results and Discussion All materials had 2n = 18 chromosomes. Chromo- somes were classified into the following six types based on the number and position of CMA-positive bands (Befu et al., 2000; Miranda et al., 1997; Yamamoto and Fig. 1. Schematic representation of chromosome types according to Tominaga, 2003; Yamamoto et al., 2007): A: two the position of CMA-positive bands. A: two telomeric and one telomeric and one proximal band, B: one telomeric and proximal band, B: one telomeric and one proximal band, C: two one proximal band, C: two telomeric bands, D: one telomeric bands, D: one telomeric band, E: without bands, Dst: telomeric band, E: without bands and Dst: type D with type D with a satellite chromosome. The gray regions indicate CMA-positive bands.

Table 1. Species belonging to the true citrus fruit trees used in this study and their CMA banding patterns of somatic chromosomes.

Latin name Distribution CMA banding patternz Clymenia polyandra (Tan.) Swing. Bismarc Archipelago 2C + 8D + 8E Eremocitrus glauca (Lindl.) Swing. Australia 2C + 9D + 7E Microcitrus australis (Planch.) Swing. Australia 1C + 11D + 6E M. australasica (F. Muell.) Swing. Australia 1B + 2C + 10D + 5E M. inodora (F. M. Bail.) Swing. Australia 8C + 7D + 2E + 1Dst M. warburgiana (F. M. Bail.) Swing. South eastern New Guinea 2A + 14D + 2Dst Sydney hybrid (M. australis × M. australasica)2C + 9D + 7E

z A: two telomeric and one proximal band, B: one telomeric and one proximal band, C: two telomeric bands, D: one telomeric band, E: without band, Dst: type D with a satellite chromosome. 26 M. Yamamoto, A.A. Abkenar, R. Matsumoto, T. Kubo and S. Tominaga

Fig. 2. CMA staining of somatic chromosomes in Clymenia, Eremocitrus and Microcitrus. 1: Cl. polyandra, 2: E. glauca, 3: M. australis, 4: M. australasica, 5: M. inodora, 6: M. warburgiana and 7: Sydney hybrid. Arrowhead indicates type D chromosome. A, B, C, D, and Dst: See Figure 1. Bar in 6 represents 5 µm for all figures. and M. warburgiana did not fall into the same category may remain that of the ancestral type. The Sydney hybrid as Citrus, Poncirus, and Fortunella. M. warburgiana is considered to be a hybrid between M. australis and possessed five characteristic chromosomes which had M. australasica (Swingle and Reece, 1967), and its a light CMA-positive band in the satellite position. female parent is probably M. australis (Asadi Abkenar M. inodora possessed eight type C chromosomes, the et al., 2004b). The chromosome configuration of the largest number in true citrus fruit trees. Sydney hybrid resembled those of M. australis and Previous studies (Asadi Abkenar et al., 2004a; M. australasica. Katayama et al., 1994; Rahman and Nito, 1994a) The chromosome configuration of Clymenia reported genetic diversity among species of Microcitrus. polyandra (2C + 8D + 8E), considered the most primi- The present study also demonstrated the variability tive of all the genera of true citrus fruit trees, and of chromosome configuration of Microcitrus. Since Eremocitrus glauca (2C + 9D + 7E), a genus very similar M. warburgiana is the only species of the genus to Microcitrus (Swingle and Reece, 1967), was quite distributed outside of Australia, the characteristics of the similar. Moreover, these chromosome configurations fruits and leaves are distinct from those of other species resembled that of M. australis (1C + 11D + 6E). Citrus of Microcitrus (Swingle and Reece, 1967). M. inodora medica is considered a primitive type of Citrus (Handa is considered to be adapted to tropical rain forests et al., 1986) and showed a 2B + 8D + 8E CMA banding (Swingle and Reece, 1967) and clearly distinguished pattern (Befu et al., 2001; Yamamoto et al., 2007). The from other Microcitrus species by the composition of CMA banding pattern of Poncirus trifoliata was 2B + essential oil in the mature leaves (Katayama et al., 1994). 10D + 6E (Miranda et al., 1997) and 4B + 8D + 6E (Befu The present findings agree with those concepts and et al., 2000). Although Clymenia, Eremocitrus, and previous findings because the chromosome configura- M. australis possessed type C chromosome instead of tions of those two species are distinct from those of other type B chromosome in C. medica and Poncirus, their species of Microcitrus and each other. Resemblance chromosome configurations were similar. These results was found in the CMA banding patterns of M. australis may indicate the existence of an ancestral type of true and M. australasica, although the latter possessed type citrus fruit trees. The chromosome configuration seems B chromosome and the former did not; however, these to have maintained the pattern of few type B or C two species were not considered very similar to each chromosomes and predominant type D and E chromo- other based on morphological traits (Swingle and Reece, somes. 1967) and DNA analysis (Asadi Abkenar et al., 2004a). The CMA banding patterns of Cl. polyandra, Differentiation of the Microcitrus species is the result M. australis, M. inodora, M. warburgiana, and Sydney of millions of years of slow evolution from the primitive hybrid were reported for the first time in this study; ancestral type (Swingle and Recce, 1967). The chromo- those of E. glauca and M. australasica have been some configuration of M. australis and M. australasica previously reported (Guerra et al., 2000). Although the J. Japan. Soc. Hort. Sci. 77 (1): 24–27. 2008. 27 high reproducibility of CMA chromosome configuration variability in Citrus species revealed by CMA/DAPI staining. is well known (Yamamoto, 2007), CMA banding Heredity 71: 234–241. patterns of these two species in the present study were Guerra, M., K. G. B. dos Santos, A. E. B. e Silva and F. Ehrendorfer. 2000. Heterochromatin banding patterns in not identical but were similar to those in studies by Rutaceae-Aurantioideae. A case of parallel chromosomal Guerra et al. (2000). There is variability within species evolution. Amer. J. Bot. 87: 735–747. at least in M. australasica (Swingle and Reece, 1967). Handa, T., Y. Ishizawa and C. Oogaki. 1986. Phylogenic study These differences may result from using different of Fraction I protein in the genus Citrus and its close related accessions in the present and previous studies. genera. Japan. J. Genet. 61: 15–24. This study demonstrated the cytogenetical character- Herrero, R., M. J. Asins, J. A. Pina, E. A. Carbonell and L. Navarro. ization of Clymenia, Eremocitrus, and Microcitrus 1996. Genetic diversity in the orange subfamily Aurantio- ideae. II. Genetic relationships among genera and species. species, because every species exhibited a characteristic Theor. Appl. Genet. 93: 1327–1334. CMA banding pattern. There seems to be a common Hizume, M. 1991. Analysis of plant chromosomes using a ground of chromosome configuration in true citrus fruit fluorescent banding method. Plant Cell Tech. 3: 78–83 (In trees; however, variability in Microcitrus chromosomes Japanese with English abstract). was also demonstrated. Further CMA staining analysis Iwamasa, M., N. Nito and J. Ling. 1988. Intra- and intergeneric using more genera and species in the subfamily hybridization in the Orange subfamily, Aurantioideae. Proc. Aurantioideae should be conducted to clarify the Sixth Int. Citrus Cong. 1: 123–130. Katayama, Y., N. Nito and J. Machino. 1994. Essential oils in phylogenic relationships among true citrus fruit trees genus Microcitrus. Bull. Fac. Agr., Saga Univ. 77: 47–51 (In and related genera. Japanese with English abstract). Literature Cited Kunitake, H., K. Haraguchi and M. Yahata. 2005. Studies on phylogenic relationship in the genus Fortunella. J. Japan. Asadi Abkenar, A., S. Isshiki and Y. Tashiro. 2004a. Phylogenic Soc. Hort. Sci. 74 (Suppl. 2): 110 (In Japanese). relationships in the “true citrus fruit trees” revealed by PCR- Miranda, M., F. Ikeda, T. Endo, T. Moriguchi and M. Omura. RFLP analysis of cpDNA. Sci. Hort. 102: 233–242. 1997. Comparative analysis on the distribution of heterochro- Asadi Abkenar, A., S. Isshiki and Y. Tashiro. 2004b. Maternal matin in Citrus, Poncirus and Fortunella chromosomes. inheritance of chroloplast DNA in intergeneric sexual hybrids Chromosome Res. 5: 86–92. of “true citrus fruit trees” revealed by PCR-RFLP analysis. Nicolosi, E., Z. N. Deng, A. Gentile, S. La Malfa, G. Continella J. Hort. Sci. Biotech. 79: 360–363. and E. Tribulato. 2000. Citrus phylogeney and genetic origin Barrett, H. C. 1985. Hybridization of Citrus and related genera. of important species as investigated by molecular markers. Fruit Variety J. 29: 11–16. Theor. Appl. Genet. 100: 1155–1166. Befu, M., A. Kitajima and K. Hasegawa. 2001. Chromosome Rahman, M. M. and N. Nito. 1994a. Inheritance and linkage composition of some citrus species and cultivars based on relationships of glutamate oxaloacetate transmianase isozyme the chromomycin A3 (CMA) banding patterns. J. Japan. Soc. in Microcitrus. J. Japan. Soc. Hort. Sci. 62: 749–754. Hort. Sci. 70: 83–88 (In Japanese with English abstract). Rahman, M. M. and N. Nito. 1994b. Phylogenic relationships Befu, M., A. Kitajima and K. Hasegawa. 2002. Classification of among the “true citrus fruit trees” by glutamate oxaloacetate the citrus chromosomes with same types of chromomycin A transmianase isozyme analysis. J. Japan. Soc. Hort. Sci. 62: banding patterns. J. Japan. Soc. Hort. Sci. 71: 394–400 (In 755–760. Japanese with English abstract). Rahman, M. M., N. Nito and S. Isshiki. 1994. Genetic analysis of Befu, M., A. Kitajima, Y. X. Ling and K. Hasegawa. 2000. phosphoglucoisomerase isozymes in ‘true citrus fruit trees’. Classification of ‘Tosa-Buntan’ pummelo (Citrus grandis [L.] Sci. Hort. 60: 17–22. Osb.), ‘Washington’ navel orange (C. sinensis [L.] Osb.) and Swingle, W. T. and P. C. Reece. 1967. The botany of Citrus and trifoliate orange (Poncirus trifoliata [L.] Raf.) chromosomes its wild relatives. p. 190–430. In: W. Reuther, H. J. Webber using young leaves. J. Japan. Soc. Hort. Sci. 69: 22–28. and L. D. Batchelor (eds.). The citrus industry, vol. 1. Carvalho, R., W. S. Soares Filho, A. C. Brasileiro-Vidal and M. University of California, Berkeley. Guerra. 2005. The relationships among lemons, limes and Yamamoto, M. 2007. Application of fluorescent staining of citron: a chromosomal comparison. Cytogenet. Genome Res. chromosomes to genetic studies in citrus. Japan. J. Plant Sci. 109: 276–282. 1: 12–19. Cornelio, W. T. M. N., A. R. S. Figueiroa, K. G. B. Santos, Yamamoto, M., A. Asadi Abkenar, R. Matsumoto, H. Nesumi, T. R. Carvalho, W. S. Soares Filho and M. Guerra. 2003. Yoshida, T. Kuniga, T. Kubo and S. Tominaga. 2007. CMA Chromosomal relationships among Citrus reticulata Blanco, banding patterns of chromosomes in major Citrus species. J. its hybrid and related species. Plant Syst. Evol. 240: 149–161. Japan. Soc. Hort. Sci. 76: 36–40. Federici, C. T., D. Q. Fang, R. W. Scora and M. L. Roose. 1998. Yamamoto, M., T. Kubo and S. Tominaga. 2005. CMA banding Phylogenic relationships within the genus Citrus (Rutaceae) patterns of chromosome of mid- and late-maturing citrus and and related genera as revealed by RFLP and RAPD analysis. acid citrus growing in Japan. J. Japan. Soc. Hort. Sci. 74: Theor. Appl. Genet. 96: 812–822. 476–478. Fukui, K. 1996. Plant chromosome at mitosis. p. 1–17. In: K. Yamamoto, M. and S. Tominaga. 2003. High chromosomal Fukui and S. Nakayama (eds.). Plant chromosome. laboratory variability of mandarin (Citrus spp.) revealed by CMA methods. CRC press, Boca Raton, Florida. banding. Euphytica 129: 267–274. Guerra, M. 1993. Cytogenetics of Rutaceae. V. High chromosomal